Oral cancer is a major health problem worldwide. Patients with early disease have better chances for cure and functional outcome, yet most patients present with advanced tumors. Early detection improves outcomes for oral cancer patients. The goal of this proposal is to develop a new technology, fluorescence spectroscopy, for non-invasive, early detection of oral cavity neoplasia and assessment of molecular changes associated with oral carcinogenesis. We will test the hypothesis that changes in biochemistry and tissue morphology produced during carcinogenesis result in alterations in the optical properties of oral mucosa. In Aim 1, we will develop diagnostic algorithms based on results of clinical trials using fluorescence and reflectance spectroscopy, and determine the sensitivity and specificity to non-invasively identify and distinguish dysplasia and early carcinoma from benign lesions and normal mucosa. We will obtain fluorescence and reflectance spectra at four different source-detector separations which sample information from different depths spanning the epithelium and superficial stroma. In Aim 2, we will investigate the biological basis for changes in fluorescence spectra of inflammatory and neoplastic lesions using short-term culture of human oral tissue. Vital microscopy using widefield Transverse slices of normal and abnormal tissue will be maintained in culture for examination with autofluorescence microscopy. We will compare the pattern of autofluorescence of normal, dysplastic and cancerous oral mucosa; changes will be related to those measured in vivo using fluorescence spectroscopy. We will explore which chromophores are responsible for oral mucosa fluorescence by comparing autofluorescence patterns to immunohistochemical patterns of fluorophores, including collagen crosslinks, NADH, FAD, cytokeratins and porphyrin, and absorbing and scattering chromophores such as hemoglobin. With this information, we will evaluate the potential of fluorescence spectroscopy as an intermediate endpoint biomarker of cancer progression. In Specific Aim 3, we will develop mathematical models to describe the fluorescence properties of oral tissue, to extract the relative contributions of principle chromophores modulated with neoplasia. We will use parameters extracted from this model to develop diagnostic algorithms based directly on alterations in tissue biochemistry and morphology that can be probed using fluorescence. Successful completion of this research program will provide a clinical tool that could dramatically improve early detection and monitoring of oral neoplasia. Technology to non-invasively assess molecular changes in oral mucosa could augment clinical and translational research of genetic mechanisms involved in carcinogenesis and treatment of oral neoplasia.